In this study, we describe the impact of genetic variation on transcript abundance in an F2 population of Arabidopsis thaliana. The RNA-seq resource generated by this study is suitable for expression quantitative trait locus (eQTL) mapping. From the aligned RNA-seq reads, and available genomic data for each of the parents of the cross, we imputed the genomes of each F2 individual (to allow genetic mapping of RNA abundance traits; briefly, genetic differences in aligned RNA-seq reads were used to impute each F2 genome). Our results show that heritable differences on gene expression can be detected using F2 populations (that is, single F2 plants), and shed light on the control of expression differences among strains of this reference plant. Overall design: 183 samples consisting of single F2 plants of a cross between Arabidopsis thaliana accessions 8230 and 6195 were generated. For each sample, RNA was collected from the aerial shoot at the 9th true leaf stage, and Illumina mRNA-seq libraries were constructed. Using these libraries, 50 bp single end RNA-seq Illumina reads were generated for each sample, and used to quantify gene expresison in each individual. The resulting expression phenotypes are suitable for genetic mapping of the control of gene expression differences in the species.